What Is a Partner Scorecard?

Every partner manager eventually faces the same awkward question: which of our partners are actually doing well? Revenue says one thing. Customer feedback says another. The partner with the biggest quarter might also be the one with the weakest certifications and the most support escalations. A partner scorecard exists to put those signals side by side so the answer doesn't depend on whoever speaks loudest in the meeting.

This article defines the partner scorecard, traces it back to the balanced scorecard and to supplier scorecards, and lists the dimensions most scorecards use. It then shows how weighting and scoring work, with an illustrative example, looks at how a large vendor program scores partners publicly, and explains how scorecards feed tiering and reviews. It closes with the mistakes that make scorecards useless.

What a Partner Scorecard Is

A partner scorecard is a periodic, weighted view of one partner's performance and capability. Three parts of that definition matter.

• Periodic. It's produced on a fixed rhythm, usually quarterly, so you can compare a partner with its own past and with its peers.
• Weighted. Several measures roll up into dimensions, and dimensions carry different weights, so the result reflects what the program values most.
• About one partner. The scorecard describes a single partner. A program dashboard describes the whole channel. You need both, but they answer different questions.

A scorecard is not a list of KPIs. A KPI catalogue is the menu of measures you could track, and partner KPIs covers that menu. The scorecard is the selection from that menu, with weights and targets attached, applied the same way to every partner of a given type. Likewise, the meeting where you discuss the result is a separate topic, covered in partner business reviews. The scorecard is the input to that conversation, not the conversation itself.

Where the Idea Comes From

The balanced scorecard

The modern scorecard traces to Robert Kaplan and David Norton's 1992 article in Harvard Business Review, The Balanced Scorecard: Measures that Drive Performance. Its opening argument is blunt: "What you measure is what you get," and a company's measurement system strongly affects the behavior of managers and employees.

Their later HBR article, Using the Balanced Scorecard as a Strategic Management System, describes the idea as supplementing traditional financial measures with criteria that measure performance from three additional perspectives: customers, internal business processes, and learning and growth. The lesson for partner management is the principle, not the four boxes. Financial results alone are a lagging view. If you only track partner revenue, you find out a partner is failing after it has already failed. For the full framework, see the balanced scorecard.

Supplier scorecards

Procurement teams have long scored suppliers on delivery, quality, responsiveness and cost, then used the scores to decide who gets more volume. A partner scorecard borrows that habit and adds something supplier scorecards often lack: growth measures. You aren't just checking that a partner behaves. You're checking whether it's building a business around your product.

The Typical Dimensions

Programs differ, but most scorecards group their measures into four dimensions. These are common patterns rather than a standard.

A few notes on each.

Revenue and pipeline is the dimension everyone expects, and it's the one most likely to dominate. Separate results from activity: closed revenue is a result, registered deals are an activity that predicts results. Mixing the two without thinking confuses a partner that's busy with one that's winning.

Capability connects to partner certification and partner enablement. It's a leading indicator. A partner with newly certified engineers hasn't yet produced revenue from them, but it's more likely to.

Customer outcomes protects you from partners who sell well and deliver poorly. It's usually the hardest to measure, because you often depend on survey data, renewal data from the partner or support records, and the data arrives late.

Engagement and compliance covers the unglamorous behaviors: showing up, reporting honestly, following the rules on deal registration and branding. Keep it small, but don't drop it. A partner that ignores process is expensive to manage even when its revenue is fine.

Some programs add a fifth dimension for strategic fit, such as alignment on target segments or investment in a dedicated team. Treat that as optional, and keep the total number of measures small enough that a partner can remember them.

Weighting and Scoring

A scorecard needs three design decisions: what's measured, how each measure becomes a score, and how scores combine.

Step 1: Pick few measures per dimension

Two or three measures per dimension is plenty. If a partner can't recite its own scorecard, it can't act on it.

Step 2: Convert each measure to a score

Raw numbers aren't comparable. Revenue is in currency, certifications are a count, satisfaction is a rating. Convert each to a common scale, such as 0 to 100, against a target. Two common methods:

• Percent of target, capped. Score equals actual divided by target, capped at 100. A partner at 80% of its revenue target scores 80.
• Banded scoring. Define bands (for example, below 50% of target scores 0, 50 to 79% scores 50, 80 to 99% scores 80, 100% or more scores 100). Bands are easier to explain and less jumpy, but they create cliffs near the edges.

Targets should differ by partner type and size. A referral partner and a systems integrator shouldn't be scored against the same revenue target. See partner onboarding for how partner types differ.

Step 3: Weight the dimensions

Weights say what the program cares about most. They must add up to 100%. A reseller-heavy program might weight revenue highest. A program that depends on implementation quality might weight customer outcomes equally. Weights should reflect strategy and change rarely.

An illustrative example

The following is entirely made up. The partner, the weights and the scores are invented to show the arithmetic, and they aren't benchmarks.

Imagine a mid-size reseller scored on a 0 to 100 scale in one quarter, using these weights:

Each weighted figure is the score multiplied by its weight (80 x 0.35 = 28.0). The total is 74.5.

The total alone hides the useful part. This partner is strong on engagement and decent on revenue, but capability is the weakest dimension. The sensible next step is an enablement plan, not a conversation about sales targets. That's the point of a scorecard: it turns a vague sense that a partner is "fine" into a specific place to act.

One refinement is worth considering. A pure weighted average lets a strong dimension mask a failing one. A partner could score 100 on revenue and 0 on compliance and still land near 70. Many programs add a floor: a minimum score on each dimension, or a minimum on any dimension flagged as critical. Microsoft's program, covered next, uses a version of this.

How Vendor Programs Score Partners in Public

Few vendors publish their internal partner scoring, but Microsoft publishes the method for its Solutions Partner designations. The partner capability score page on Microsoft Learn is a rare public example of a weighted partner scorecard.

According to that page:

• The partner capability score is a composite score across three categories: Performance, Skilling and Customer success.
• It's calculated from information already captured in Partner Center.
• Each of six solution areas has a maximum of 100 points across the three categories. The six are Data & AI (Azure), Infrastructure (Azure), Digital & App Innovation (Azure), Business Applications, Modern Work and Security.
• To qualify for a Solutions Partner designation, a partner needs a score of at least 70 points for the solution area, and every metric in the area must be greater than zero points.

The categories map neatly to the dimensions above:

Several design choices on that page are worth copying in spirit.

• A pass mark plus a floor. The 70-point threshold sets the bar, and the rule that every metric must exceed zero stops a partner from passing on one strong category alone.
• Partial credit. Reaching a metric's threshold earns the maximum weighted points for it, and partial progress earns partial points.
• Different tracks for different partners. For several solution areas, Microsoft describes separate Enterprise and small and medium business paths with their own thresholds.
• Net, not gross. Net customer adds subtracts lost customers from new ones across the trailing 12 months, so churn counts against the partner.
• Visible data and a refresh rhythm. Partners can see their score and download the data behind it. The page says performance and customer success data is typically refreshed by the 20th of each month, and skilling data within about 10 days of a certification being completed.

Note what the page doesn't give. Its summary table shows the scoring framework with placeholder values rather than the points per metric, so don't assume a split between categories from it. For actual weights, a partner has to consult the requirements for the relevant solution area.

Microsoft's scale is far beyond what most vendors need. The lesson is the structure: a small number of categories, a pass mark, a floor, transparent data and a predictable refresh cycle.

How Scorecards Feed Tiering and Reviews

Once you have a consistent score, it plugs into three decisions.

Tiering. Programs often group partners into tiers (for example registered, select and premier) with different benefits. A scorecard gives tiering an evidence base. Rather than tiering on revenue alone, you can set tier thresholds on the composite score, with a minimum in certain dimensions. Microsoft's designations work this way, with a score threshold gating a status. Tiers and benefits are usually paired with partner incentives, so a score that moves a partner up a tier should have a clear effect on margin, rebates or access.

Business reviews. The scorecard is the opening page of a review. It shows trends since the last period, names the weakest dimension and sets the agenda. See partner business reviews for how to run that meeting.

Resource allocation. Scores tell you where to spend enablement time, co-marketing budget and partner manager attention. A low-capability, high-engagement partner is a good enablement investment. A high-revenue, low-compliance partner needs a different conversation.

Key Facts: Partner Scorecard

• A partner scorecard is a periodic, weighted view of one partner's performance and capability, built from a small set of dimensions with targets and weights.
• Kaplan and Norton's 1992 HBR article introduced the balanced scorecard, built on the argument that "what you measure is what you get" (HBR).
• Microsoft's partner capability score covers three categories (Performance, Skilling, Customer success) with a maximum of 100 points per solution area (Microsoft Learn).
• To qualify for a Solutions Partner designation, a partner needs at least 70 points in the solution area and more than zero points on every metric (Microsoft Learn).
• Microsoft's net customer adds metric counts new eligible customers in the trailing 12 months and subtracts lost customers (Microsoft Learn).
• Typical dimensions are revenue and pipeline, capability, customer outcomes, and engagement and compliance.

Common Pitfalls

• Too many measures. A 30-line scorecard hides what matters. Partners ignore it and managers can't explain it.
• Measuring only what's easy. Revenue is easy to pull. Customer outcomes are hard, so they get skipped, and the scorecard drifts back to a revenue report.
• One scorecard for every partner type. A referral partner, a reseller and an integrator do different jobs. Use a shared structure with type-specific targets.
• Secret scoring. If partners can't see how they're scored, they can't improve, and they'll suspect the numbers are political. Share the method and the data.
• Weights that never change. Priorities move. Review weights once a year, and don't change them mid-period.
• Gaming. Any target gets gamed. A partner that's scored on registered deals may register junk deals. Pair activity measures with result measures so gaming one doesn't help.
• Scores with no consequence. If a score never changes tier, benefits or attention, it's decoration. Decide in advance what a low or high score triggers.
• Averages that hide failure. Use floors so a strong dimension can't cover a failing one.
• Stale data. A score built on figures three quarters old describes a different partner. Fix the refresh rhythm and stick to it.

For the wider setting, see partner-led growth, and for how scorecard measures relate to company-wide goal setting, see KPIs.

Related Reading

• Partner KPIs
• Partner business reviews
• Partner certification programs
• Partner enablement
• Partner incentives: margins, rebates and SPIFFs
• The balanced scorecard
• Partner-led growth
• KPIs